Sai Surya Duvvuri

The Art of Scaling Reinforcement Learning Compute for LLMs

LUCID: Attention with Preconditioned Representations

Let's (not) just put things in Context: Test-time Training for Long-context LLMs

Fast and Simplex: 2-Simplicial Attention in Triton

LoRA Done RITE: Robust Invariant Transformation Equilibration for LoRA Optimization

LASER: Attention using Exponential Transformation

LASER: Attention with Exponential Transformation